MEMORANDUM
PER CURIAM.The complaint in this class action was filed March 8, 1978, to challenge the constitutionality of the apportionment of State Senatorial Districts enacted by the Tennessee General Assembly in 1973. This court, on January 15, 1979, held that apportionment plan to be unconstitutional because it allowed a gross maximum deviation from population equality of 18.03 per cent. We enjoined defendants from conducting any further elections under said plan. We retained jurisdiction to impose a plan of apportionment if the General Assembly failed to enact a constitutional reapportionment plan by June 1, 1979. The January 15, 1979, opinion of this court, heretofore unpublished, is made Appendix A to this opinion. Prior to the June 1, 1979, deadline, the legislature passed Chapter 443 of the Acts of 1979, now codified as Tennessee Code Annotated § 3-1-102, which reapportioned the Senatorial Districts. Plaintiffs thereupon filed a motion for further relief. (Note: the number of Code § 3-102 cited in Appendix A has been changed to § 3-1-102.)
Meanwhile, defendants appealed the January 15 order of this court to the Supreme Court of the United States, which held, in light of the later reapportionment act adopted by the legislature, that the issues raised by the State were moot. On October 1, 1979, the Supreme Court entered an order directing that the judgment of the district court be vacated and that the entire action be dismissed. Thereafter, in a per curiam opinion announced February 19, 1980, the Supreme Court granted a petition for rehearing, vacated the order of October 1, 1979, and remanded the cause to this court for “such further proceedings . as may be appropriate.” See Crowell v. Mader, No. 78-1780, 444 U.S. 505, 100 S.Ct. 992, 62 L.Ed.2d 701. The case now is before the court for consideration of plaintiffs’ motion for further relief.
The parties have stipulated that, under the present senatorial district apportionment provided by Tennessee Code Annotated § 3-1-102, the gross maximum deviation from the ideal of population is .89 percent.1 The stipulated population figures for the Senatorial Districts is made Appendix B to this opinion. Plaintiffs advance no claim that the State has violated the equal protection clause by failing to construct the districts “as nearly of equal population as is practicable.” Reynolds v. Sims, 377 U.S. 533, 577, 84 S.Ct. 1362, 1390, 12 L.Ed.2d 506, 536 (1964).
Plaintiffs challenge the statute on two other grounds, however. First, they contend that two of the districts lack internal contiguity, in that a part of each district is not' contiguous with the remainder of the same district. Second, they complain that the statute causes the disenfranchisement, for purposes of electing State Senators, of voters who have been shifted from even-numbered to odd-numbered districts. After reviewing the record and hearing oral argument by counsel, the court concludes that plaintiffs are not entitled to further relief on either of these grounds.
I
Plaintiffs assert that Districts Nine and Ten lack internal contiguity.
District Nine contains the northern portion of Hamilton County, most of Meigs County, and all of several other counties. The Hamilton County portion of District Nine touches the remainder of the district only at one portion of the common bound*228ary between Hamilton and Meigs Counties; at that point the two counties are divided by the Tennessee River. No bridge or ferry crosses the river within the district.
In District Ten, all of which is within Hamilton County, the Airport and Eastdale precincts are nearly cut off from the remainder of the district by the Missionary Ridge precinct of another district. A narrow neck of land connects the two portions, but no road runs over this land. Plaintiffs contend that this alleged lack of contiguity violates federal and state law.
The federal constitution does not require either congressional or state legislative districts to be contiguous. See, e. g., Wood v. Broom, 287 U.S. 1, 53 S.Ct. 1, 77 L.Ed. 131 (1932); Ferrell v. Oklahoma, 339 F.Supp. 73, 82 (W.D.Okl.), aff’d mem. sub nom. Ferrell v. Hall, 406 U.S. 939, 92 S.Ct. 2045, 32 L.Ed.2d 328 (1972); Skolnick v. State Election Board, 336 F.Supp. 839 (N.D. Ill.1971); Ince v. Rockefeller, 290 F.Supp. 878, 883 (S.D.N.Y.1968). See also Wright v. Rockefeller, 376 U.S. 52, 59 n. 1, 84 S.Ct. 603, 607 n. 1, 11 L.Ed.2d 512, 517 (1964) (Douglas, J., dissenting).
Plaintiffs concede this, but they argue that federal courts have expressed a preference that legislative districts be contiguous. However, the United States Supreme Court decisions cited by plaintiffs do not even intimate that contiguity is a preferred redistricting goal. They suggest only the limited principle that a state may justify minor deviation from population equality in order to “provide for compact districts of contiguous territory . . ..” Reynolds v. Sims, supra, 377 U.S. at 578, 84 S.Ct. at 1390, 12 L.Ed.2d at 537. See, e. g., Chapman v. Meier, 420 U.S. 1, 23, 95 S.Ct. 751, 764, 42 L.Ed.2d 766, 782 (1975); Swann v. Adams, 385 U.S. 440, 444, 87 S.Ct. 569, 572, 17 L.Ed.2d 501, 504 (1967).2 Federal law, therefore, neither requires nor expresses a preference for contiguity.
Plaintiffs next contend that the Tennessee Constitution requires contiguity, and that this court should modify the boundaries of Districts Nine and Ten on that ground. The Constitution of Tennessee, Article 2, Section 6, states, in pertinent part, that “[i]n a district composed of two or more counties, each county shall adjoin at least one other county of such district . .” While this provision clearly has no application to District Ten which is contained in a single county, it may impose a requirement of contiguity on District Nine. Defendants argue that even if such a requirement exists, the District Nine portions of Hamilton County and Meigs County do adjoin each other so that there is no lack of contiguity.3
This question, of course, involves no federal constitutional claim and, therefore, is not within the court’s jurisdiction under 28 U.S.C. § 2284(a). The court concludes, however, that since this claim and the federal claims asserted by plaintiffs in this case are derived from a “common nucleus of operative fact,” and since the federal claims asserted are of a “substance sufficient to confer subject matter jurisdiction on the court,” this State law question falls within the court’s pendent jurisdiction. Sullivan v. Crowell, 444 F.Supp. 606 (W.D.Tenn.1978). See generally United Mine Workers v. Gibbs, 383 U.S. 715, 86 S.Ct. 1130, 16 *229L.Ed.2d 218 (1966); Seals v. Quarterly County Court, 526 F.2d 216 (6th Cir. 1975).
Contiguity, in the sense that each part of a whole adjoins 4 another part of the whole, is solely a territorial concept. Contiguity is absent, then, only when a portion of a district is separated from the remainder of the district by the intervention of the territory of another district. Under the Tennessee Constitution, a county “adjoins” another county that is part of the same district unless another district separates the two counties. Therefore, the mere fact that the Tennessee River separates the District Nine portion of Hamilton County from another county in the district does not mean that the counties do not adjoin or are not contiguous.
According to plaintiffs, however, the lack of a bridge or ferry traversing the river from the Meigs County bank to the Hamilton County side renders District Nine non-contiguous. They argue that “[a] contiguous district is one in which a person can go from any point within the district to any other point without leaving the district.” Comment, Reapportionment, 79 Harv.L. Rev. 1228 (1966) (no authority cited).
This definition per se is not inconsistent with the court’s view that District Nine is contiguous. No particular mode of travel is specified. A person obviously could cross the river by boat without entering another district. Plaintiffs’ reading, however, requires an inference that only terrestrial, as distinguished from marine, forms of transportation are intended. The court does not believe that convenience or ease of travel is an essential element of contiguity. Plaintiffs’ proposed limitation would distort an otherwise plain and functional definition.
Plaintiffs further argue that contiguous territory is “territory touching, adjoining, and connected, as distinguished from territory separated by other territory,” In re Sherrill, 81 N.E. 124, 131, 188 N.Y. 185 (1907). They contend that the Tennessee River, which separates Meigs County from Hamilton County, constitutes “other territory” which renders the district noncontiguous. In In re Sherrill, supra, the court held that Richmond County, New York, an island in the Atlantic Ocean, was not contiguous with Queens County on the mainland, because the water separating the two was “other territory.” There is an obvious difference between the Tennessee River and the body of water involved in Sherrill.
This court declines to adopt any definition that would treat the Tennessee River as other territory.5 The Supreme Court of Michigan dealt with a situation somewhat similar to the present litigation in Board of Supervisors v. Blacker, 92 Mich. 638, 646, 52 N.W. 951, 953 (1892), a legislative reapportionment case. The court said:
Some argument is made that the Legislature was bound under the provisions of section 3, art. 4, declaring that “each representative district shall consist * * * of convenient and contiguous territory,” to unite Keweenaw and Isle Royal counties to the county of Houghton, for the reason that, within the meaning of the Constitution, they were not convenient and contiguous territory to any other county. This clause in the constitution does not bear the restricted meaning contended for. It does not mean in contact by land. Certainly, so far as the islands are concerned, they may be considered contiguous, although separated by wide reaches of navigable deep waters. (Emphasis supplied.)
We believe that a district lacks contiguity only when a part is isolated from the rest by the territory of another district. Therefore, District Nine is contiguous and adjoin*230ing within the meaning of Article 2, § 6, of the Constitution of Tennessee, and plaintiffs’ claim for further relief on this ground must fail.
II
Plaintiffs also challenge the statute on the ground that it unconstitutionally disenfranchises voters who otherwise could participate in the 1980 senatorial elections. Under Article 2, § 3, of the Tennessee Constitution, senators are elected for staggered four-year terms, and voters in even-numbered districts will be electing senators in 1980.
Prior to 1966, members of the Tennessee State Senate, like members of the House of Representatives, were elected for two-year terms. By an amendment to Article 2, § 3, of the State Constitution adopted in 1966, the terms of Senators were changed to four years on a staggered basis. Section 3 provided orderly transition to the new staggered system by requiring that in the first election under the new amendment Senators elected, in districts designated by even numbers would be elected for four years and those in districts designated by odd numbers would be elected for two years. Thereafter, Senators in both odd-numbered and even-numbered districts would be elected for four years.
The parties have stipulated that the challenged reapportionment plan shifts approximately 117,000 persons from even-numbered districts to odd numbered. The stipulation is that 117,000 persons have been shifted. The number of voters in the group is unknown. Such shifted persons, who last could have voted for State Senators in 1976 and who otherwise would be eligible to vote in 1980, will not be entitled to vote for a State Senator again until 1982.
Defendants insist, and plaintiffs concede, that such shifting of voters is a natural and mathematically inevitable byproduct of reapportionment. Plaintiffs further concede that, had this particular instance of voter shifting resulted from a reapportionment plan adopted in response to a decennial census, no constitutional challenge would be possible. Plaintiffs contend, however, that since the instant reapportionment was ordered by this court to correct an unconstitutional apportionment plan enacted earlier in the decade, the resulting disenfranchisement cannot be tolerated. The remedy plaintiffs seek is to require defendants to conduct elections for all thirty-three Senate seats in 1980.
Plaintiffs’ argument depends on the premise that the legislature’s prior unconstitutional apportionment plans taint the otherwise unobjectionable plan we now consider. The present plan, with its gross deviation of less than one per cent from perfect mathematical equality, conforms to the Supreme Court’s mandate to create districts that are “as nearly of equal population as is practicable.” Reynolds v. Sims, supra, 377 U.S. at 577, 84 S.Ct. at 1390. Nor do plaintiffs allege that the General Assembly intended to discriminate in developing the reapportionment plan. Indeed, the plaintiffs readily acknowledge that some odd-even shifting of voters is inevitable whenever Senatorial Districts are reapportioned. Plaintiffs’ sole objection to the present plan is that it was necessitated by the holding of this court that its predecessor plan was unconstitutional.
We think the plaintiffs’ argument is without merit. In the first place, the argument necessarily implies that the General Assembly was without power to enact a constitutionally valid apportionment plan unless it violated the staggered terms requirement of the Tennessee Constitution. We decline so to hold. In the second place, at least two other courts have found no constitutional violation where some voters, as a byproduct of court ordered reapportionment, were disenfranchised temporarily by being shifted from one district to another. See Pate v. El Paso County, 337 F.Supp. 95 (W.D.Tex.), aff’d without opinion, 400 U.S. 806, 91 S.Ct. 55, 27 L.Ed.2d 38 (1970); Carr v. Brazoria County, Texas, 341 *231F.Supp. 155 (S.D.Tex.), aff’d, 468 F.2d 950 (5th Cir. 1972).
Furthermore, plaintiffs’ proposed remedy, ordering all State Senators to stand for reeleetion in 1980 despite the State Constitution’s provision to the contrary, is not supported by precedent. The only case cited by plaintiffs imposing a comparable remedy is Chavis v. Whitcomb, 307 F.Supp. 1362, 1367 (S.D.Ind.1969), revd. on other grounds, 403 U.S. 124, 91 S.Ct. 1858, 29 L.Ed.2d 363 (1970). The Chavis court ordered all Indiana state senators to stand for reelection as part of its judicially developed plan to render all districts single member. “To permit the present terms of incumbent senators to extend beyond the 1970 general election would,” the court said, “prevent implementation of the districting plan herein prescribed.” 307 F.Supp. at 1367. The court noted that “The Indiana constitutional provision for staggering the terms of senators ... is entirely proper and valid and would be mandatory in a legislatively devised redistricting plan.” Id. Furthermore, the mandate of the court in Chavis was never carried out because the Supreme Court reversed the finding of an underlying constitutional violation. Faced as we are with a legislatively devised reapportionment plan that achieves near perfect mathematical equality among the districts, we find Chavis unpersuasive as authority for the remedy advocated by plaintiffs.
This court is of the opinion that plaintiffs’ claims fail to rise to constitutional magnitude. The temporary disenfranchisement of these voters violates neither the equal protection clause nor any other constitutional provision. See, e. g., Ferrell v. Oklahoma, 339 F.Supp. 73, 82 (W.D.Okl.), aff’d mem. sub nom. Ferrell v. Hall, 406 U.S. 939, 92 S.Ct. 2045, 32 L.Ed.2d 328 (1972); Stout v. Bottorff, 249 F.Supp. 488, 495 (S.D.Ind.1965). Shifts from odd-numbered to even-numbered districts and vice versa are an unavoidable consequence of the reapportionment ordered by this court.
Moreover, the deprivation suffered is de minimis at most and the remedy urged by plaintiffs would not justify the massive intrusion into the state’s political machinery urged by plaintiffs. The disenfranchisement is temporary in nature and is no different from that experienced by “new registrants who reach the age of 18 years shortly after an election and [by] people moving from one area to another.” Ferrell v. Oklahoma, supra, 339 F.Supp. at 82. Furthermore, the court believes that politically sensitive senators will seek to represent effectively those shifted voters who likely will be part of their constituencies in the 1982 election.
Plaintiffs submitted no evidence that the General Assembly made these shifts for invidious or discriminatory purposes. Rather, this disenfranchisement results simply from the neutral and inoffensive concatenation of the Tennessee Constitutional provision for overlapping senatorial terms, this court’s order requiring reapportionment, and the legislature’s laudable objective to achieve near perfection in equalizing the population of senatorial districts. Accordingly, plaintiffs claim on this ground cannot prevail.
It is significant that the third prayer of the complaint in this case, filed March 8, 1978, was as follows: “That this Honorable Court order elections to be conducted in all thirty-three (33) Senate Districts in 1978 pursuant to said proper plan of apportionment adopted by this court.” This court did not adopt its own apportionment plan, as prayed by plaintiffs. Instead we held the existing plan to be unconstitutional and allowed time for the General Assembly to adopt a plan meeting the one person, one vote standard. In our former opinion (Appendix A), we did not grant the prayer of the complaint that all thirty-three State Senators be reelected at the same time, although it was inevitable that shifts would have to be made between odd-numbered and even-numbered Senatorial Districts in achieving a valid plan. Having not granted this extraordinary relief in our earlier opinion (Appendix A), we respectfully decline to do so now.
*232
Before PHILLIPS, Chief Circuit Judge, BROWN, Chief District Judge, and MORTON, Chief District Judge.PER CURIAM.
Plaintiffs in this class action are Tennessee voters seeking reapportionment of the state senatorial districts, which they allege are unconstitutionally apportioned pursuant to Tennessee Code Annotated section 3-1-102. The suit is brought pursuant to 42 U.S.C. §§ 1983 and 1988, and jurisdiction is conferred upon this court by 28 U.S.C. § 1343(3). A panel of three judges was designated pursuant to 28 U.S.C. § 2284. The court finds that this cause meets the requirements for a class action set out in Rule 23 of the Federal Rules of Civil Procedure.
Plaintiffs assert that the present plan of apportionment of state senatorial districts violates the one person, one vote mandate of Baker v. Carr, 369 U.S. 186, 82 S.Ct. 691, 7 L.Ed.2d 663 (1962), and Reynolds v. Sims, 377 U.S. 533, 84 S.Ct. 1362, 1392, 12 L.Ed.2d 506 (1964). This plan was enacted by the Tennessee General Assembly in 1973 after two prior plans (a primary plan and an alternate plan) had been declared unconstitutional in Kopald v. Carr, 343 F.Supp. 51 (M.D.Tenn.1972). Because there was not time after the Kopald decision for the legislature to enact a new plan before the 1972 elections, the court ordered that its own plan, a modified version of the General Assembly’s alternate version, be used for those elections only. The court recognized, however, that legislative apportionment is primarily and properly a legislative function, id. at 53, and stated:
If the General Assembly enacts a plan of reapportionment prior to July 1, 1973, and such plan is not challenged in this case within thirty (30) days after the signing thereof by the Governor, this case will be closed. If, however, no plan is enacted, or, if enacted, is successfully challenged, the court will formulate a reapportionment plan.
Id. at 54. As indicated above, the General Assembly responded by enacting the plan that is basically still in effect.1
Counsel for defendants argues that plaintiffs are barred by laches, asserting that the Kopald court “stated that an attack on the reapportionment would be barred if plaintiffs did not challenge [the General Assembly’s] plan within 30 days,” Defendants Reply Brief and Argument, filed November 3, 1978, at 3. This overstates the Kopald decision, which only provided that that case would be closed if no challenge promptly appeared. The present individual plaintiffs were not registered voters at that time and so were not members of the Kopald plaintiff class. It is stipulated that each of the individual plaintiffs registered to vote in Tennessee for the first time in 1975 or later. Each was only eligible to vote in one major election before filing this suit, and thus plaintiffs are not guilty of laches. Any considerations of finality of judgment or of the desirability of maintaining a stable apportionment scheme are outweighed by the importance of plaintiffs’ rights to vote and to have their votes count equally with those of other voters throughout the state.
*233The following facts pertinent to the merits of this ease are stipulated:
(1) Population figures are based on the 1970 Federal Decennial Census.
(2) 1970 census population figures for voting precincts are interpolations prepared by the Office of Legal Services of the Tennessee General Assembly and derived in part from local authorities.
(3) The optimum population per senatorial district based on the 1970 census is 118,-914 (total Tennessee population of 3,924,164 divided by 33 districts).
(4) The gross maximum deviation from the optimum under Tennessee Code Annotated section 3-102 is 18.03 percent, being the total of the greatest positive deviation, + 10.47 percent (Senate District 4, population 131,356), and the greatest negative deviation, -7.56 percent (Senate District 19, population 109,925).2
It is the opinion of this court that Tennessee Code Annotated section 3-102 is unconstitutional because it does not “construct districts ... as nearly of equal population as is practicable.” Reynolds v. Sims, supra, 377 U.S. at 577, 84 S.Ct. at 1390. The practicality of smaller deviations was demonstrated when the 1972 elections were held pursuant to the Kopald plan, under which the variance from the optimum was below four percent. See Kopald v. Carr, supra, 343 F.Supp. at 53. It is true that sizable deviations have been allowed where such variations are justified by “legitimate considerations incident to the effectuation of a rational state policy.” See, e. g., Mahan v. Howell, 410 U.S. 315, 325, 93 S.Ct. 979, 985, 35 L.Ed.2d 320 (1973) (quoting Reynolds v. Sims, supra, 377 U.S. at 579, 84 S.Ct. at 1390). In Mahan, the court upheld a deviation of 16.4 percent where it was found to be in furtherance of the state’s “policy of maintaining the integrity of political subdivision lines.” Id., 410 U.S. at 325, 93 S.Ct. at 985.
It is clear that the burden is on the state to prove a policy justification for the deviations complained of in the instant case. In Wells v. Rockefeller, 394 U.S. 542, 89 S.Ct. 1234, 22 L.Ed.2d 535 (1969), the Supreme Court held that the burden is on the proponent of a districting plan to justify deviations. Id., 394 U.S. at 544, 89 S.Ct. at 1236. Concededly, this principle was significantly qualified in Gaffney v. Cummings, 412 U.S. 735, 93 S.Ct. 2321, 37 L.Ed.2d 298 (1973), which emphasized that stricter standards must be applied to congressional districting such as was involved in Wells than to districting for state legislative bodies, and which held that “minor deviations from mathematical equality among state legislative districts are insufficient to make out a prima facie case of invidious discrimination under the Fourteenth Amendment so as to require justification by the State.” Id., 412 U.S. at 745, 93 S.Ct. at 2327. The Gaffney Court indicated that deviations from the optimum district size may be categorized into three groups: those “too great to be justified by any state interest,” “minor deviations” not calling for state justification, and, most significant for the case at bar, deviations “sufficiently large to require justification.” Id. The case cited by the court as an example of this intermediate group *234was Mahan v. Howell, supra, with its maximum deviation of 16.4 percent. The finding in Mahan was based on facts offered by the state, and the court noted that in Swann v. Adams, 385 U.S. 440, 87 S.Ct. 569, 17 L.Ed.2d 501 (1967), a large deviation was disapproved where the state “offered no evidence at the trial level to support the challenged variations . . Mahan, supra, 410 U.S. at 328, 93 S.Ct. at 987.
The 18.03 percent variance now at issue certainly falls in the Mahan category, yet in this case, unlike Mahan, no justification has been proved by the state. Although defendants point out that Article 2, section 6 of the Tennessee Constitution “prefers districts that contain whole contiguous counties,” (Defendants’ Reply Brief and Argument, filed November 3,1978, at 7), defendants have failed to indicate how the plan under attack furthers this preference or even that the preference rises to the level of an established state policy. Tennessee Code Annotated section 3-1-102 creates a number of districts that cut across county lines, and several of these districts deviate markedly from the optimum. Although Mahan teaches that other policy considerations might justify exceptions to a general policy of observing existing political boundaries, no such justifications have been identified for the noncontiguous districts now existing in this state.
It having been demonstrated that Tennessee senatorial districts with substantially smaller deviations from the optimum are practicable and that the magnitude of the variations created by the present plan is without justification, this court holds that Tennessee Code Annotated section 3-1-102 is unconstitutional. Defendants are enjoined from conducting or causing to be conducted any primary or general election of state senators pursuant to its provisions. The court will retain jurisdiction of this cause, and if the Tennessee General Assembly has not enacted a constitutional scheme of apportionment of state senatorial districts prior to June 1, 1979, the court will reinstate the Kopald plan 3 or will effectuate such other plan as may be presented to the court that surpasses the Kopald plan in achieving the ideal of one person, one vote.
Plaintiffs seek attorney fees pursuant to 42 U.S.C. § 1988, as amended by the Civil Rights Attorney’s Fee Awards Act of 1976, which provides in pertinent part:
In any action or proceeding to enforce a provision of sections 1981, 1982, 1983, 1985, and 1986 of this title, . . the court, in its discretion, may allow the prevailing party, other than the United States, a reasonable attorney’s fee as part of the costs.
It is the opinion of this court that such a fee should be awarded in this case. A separate hearing will be held to determine the amount to be so awarded.
An appropriate order will be entered.
ORDER AMENDING MEMORANDUM
In accordance with the Amendment to Stipulation filed by the parties to this cause on May 1,1979, it is hereby ORDERED that this Court’s memorandum opinion entered January 15, 1979, be amended as follows: In footnote 2 of said memorandum, the population listed for Senate District 14 shall be changed from 118,941 to 118,041, and the population listed for Senate District 15 shall be changed from 127,329 to 123,262.
STIPULATION OF POPULATION
Pursuant to the Federal Rules of Evidence, plaintiffs and defendants, for the purpose of this action, do hereby stipulate and agree to the following:
*2351. Population data is based upon. the 1970 Federal Decennial Census, as corrected.
2. The population figures for Tennessee’s thirty-three (33) state senatorial districts, as composed under T.C.A. § 3-1-102 (1979), with the exception of senatorial districts 10 and 11 and that portion of senatorial district 9 within Hamilton County, Tennessee, were prepared by the Office of Local Government, Comptroller of the Treasury, in accordance with the procedure described in the affidavit of Frank D. Hinton, Director of the Office of Local Government, which is attached hereto as Exhibit A and incorporated by reference into this stipulation.
3. The corrected optimum population per senatorial district, based upon corrected 1970 census data, is 118,970.
4. The population, its variance in numbers of people from the optimum district, and its percentage of variation from the optimum districts for each Tennessee senatorial district enumerated below, as composed under T.C.A. § 3-1-102 (1979), is as follows:
5. The population figures for state senatorial districts 10 and 11 and that portion of senatorial district 9 within Hamilton County, Tennessee, as composed under T.C.A. § 3-1-102 (1979), were prepared by Dr. Richard Lee Wilson, Registrar-at-Large for Hamilton County, Tennessee, and were based on 1970 census data and were compiled in accordance with the procedure described in the depositions of Dr. Richard Lee Wilson taken on September 25, 1979, and March 7, 1980 and filed with the Court on October 10, 1979 and March 13, 1980, respectively.
6. The population of senatorial districts 10 and 11, as composed under T.C.A. § 3-1-102 (1979), its variance in number of people from the optimum district, and its percentage of variation from the optimum district, is as follows:
7. The population of that portion of senatorial district 9 within Hamilton County, Tennessee, as composed under T.C.A. § 3-1-102 (1979), is 16,492. The total population of senatorial district 9, its variance in number of people from the optimum district and its percentage of variation from the optimum district is as follows:
8. Of Tennessee’s thirty-three (33) senatorial districts, as composed under T.C.A. § 3-1-102 (1979), the greatest positive deviation from optimum district size is +522 people, which equals a percentage deviation of + .44%, and the greatest negative deviation from optimum district size is -533 people, which equals a percentage deviation of -.45%. Combining the greatest positive deviation from optimum district size with the greatest negative deviation yields a total gross deviation of .89%.
Approved by counsel for the parties, this 13th day of March, 1980.
(si John L. Ryder JOHN L. RYDER Attorney for Plaintiffs
(s) William W. Hunt. III WILLIAM W. HUNT, III Assistant Attorney General
(s) Robert B. Littleton ROBERT B. LITTLETON Deputy Attorney General
*236Attorneys for Defendants
OF COUNSEL:
WILLIAM M. LEECH, JR. Attorney General
State of Tennessee
450 James Robertson Parkway
Nashville, TN 37219
AFFIDAVIT
FRANK D. HINTON, being duly sworn according to law, deposes and states as follows:
1. I am the Director of the Office of Local Government, a division of the office of the Comptroller of the Treasury for the State of Tennessee.
2. In March of 1979 the Office of Local Government was requested to provide staff assistance to a Senate Subcommittee charged with the responsibility of drafting an acceptable apportionment plan for the Tennessee Senate. A decision was made by the Committee to draw districts with the smallest possible variation from the average district size of 118,970 persons. A decision was also made not to place two then serving senators in the same district. It was agreed that a number of counties would split between two districts but that we should avoid placing a single county in three districts. Present district numbers were to be used where possible.
3. The procedure used in apportionment is quite simple in theory but may be somewhat more complex in application. Since the 1970 U.S. Census data provides the only true statewide picture of population, it was chosen as the base for all calculations. Maps of most counties showing Census Enumeration Districts (ED’s) and County Legislative Districts (MD’s) were used to locate more limited population groups for allocation to specific districts.
4. Beginning with District I and continuing through the remaining thirty-two districts, counties were aggregated until persons included approached the desired 118,-970. If the number proved to be short, a portion of an adjoining county was added. If the number was too large, a portion of one county was moved to an adjoining district.
5. In dividing counties, locally created districts or precincts were used in most instances. Since split precincts make voting more difficult, an effort was made to avoid the creation of such situations. In many instances, local political districts have little relationship to ED’s. This necessitates the determination of number of persons in a district by a procedure which we call “House Count.” House Count involves actually counting the number of houses shown on County highway map of the affected area. Houses are counted in each portion of any ED within each area to be divided. Next, each house in each affected ED is counted. These totals are then compared with the number of houses as determined by the Census. Since most maps were prepared somewhat prior to the time of taking the Census, house count is usually low and must be adjusted. This may be accomplished either by percentage or ratio. After determining the adjusted number of houses in each portion of an area to be divided, house count is multiplied by the average residents per house as shown by Census data. The appropriate number is then allocated to each senate district.
6. In some Counties, particularly in the larger ones, data is available on a block basis and therefore accuracy is much more certain. Where population is determined by house count, it is reasonable to concede that some slight error may be present.
AND FURTHER AFFIANT SAITH NOT.
Is) Frank D. Hinton FRANK D. HINTON
*237Sworn to and subscribed before me this 13th day of-March, 1980.
(s') Patricia M. Mitchell NOTARY PUBLIC
My Commission Expires: December 8, 1982
MEMORANDUM
. The greatest positive deviation is .44 per cent in District Eight and the greatest negative deviation is .45 per cent in District Nineteen.
. The teaching of Connor v. Finch, 431 U.S. 407, 97 S.Ct. 1828, 52 L.Ed.2d 465 (1977), is not inconsistent with this view. There the district court, in developing a reapportionment plan, stated that maintaining contiguity would be one of its guidelines. Departures therefrom in the drawing of district lines gave rise to complaints that black voting strength was being diluted impermissibly. The Supreme Court required that the district court redraw the lines consistent with the guideline of achieving contiguity and compactness so that suspicions of racial discrimination would be put to rest.
. The court believes that the term “adjoin” in the Tennessee Constitution and the term “contiguous” have essentially identical meanings. In fact, the dictionary definition of “contiguous” is "next or adjoining with nothing similar intervening . . Webster’s Third New International Dictionary 492 (Miriam Unabridged ed. 1966) (emphasis added).
. The term comes from Tennessee Const, art. 2, § 6, supra.
. Under a contrary holding, both Districts Twenty-three and Twenty-six would lack contiguity because each contains a county that is separated by the Tennessee River from another county in the district. District Twenty-seven,, containing land that is on the Arkansas side of the Mississippi River, would suffer a similar infirmity.
. With the exception of the so-called “Gillock Amendment,” which was held unconstitutional in Brawner v. Crowell, 434 F.Supp. 1119 (1977), amendments to the plan since 1973 were enacted when precinct configurations or designations were altered, necessitating changes in the descriptions of senatorial districts in order to maintain the district lines as originally enacted.
. The following table, stipulated in this case, lists each senate district and its percentage deviation from the optimum:
. If it becomes necessary for the court to impose the Kopald plan, any modifications required by changes in the configurations or designations of voting precincts used to define the districts in that plan will, of course, be made.
As corrected by order entered May 15, 1979, pursuant to stipulation of the parties